© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 2-MLP-SM
    11. 24213
    Prev
    Next
    INDEX
    Explanations

    instances of the word "provide" and related terms indicating the act of supplying or offering something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/2-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.2.hook_mlp_out
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    255
    +0.11
    0.6%
    21
    +0.11
    0.6%
    369
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    255
    +0.11
    0.06
    412
    +0.11
    0.05
    492
    +0.10
    0.04
    Negative Logits
    poons
    -1.71
     sense
    -1.54
     Eqs
    -1.50
     heter
    -1.50
     shapes
    -1.47
     flesh
    -1.44
     simplicity
    -1.40
     decom
    -1.38
     Eq
    -1.37
     Shape
    -1.36
    POSITIVE LOGITS
    stre
    1.55
    opinion
    1.53
    erta
    1.50
    izumab
    1.49
    iterr
    1.44
    NOTES
    1.44
    jours
    1.44
    reader
    1.43
    aucoup
    1.43
    ')->
    1.43
    Activations Density 0.218%

    No Known Activations